Trends in memory function and memory impairment among older adults in the USA and Europe, 1996-2018.

Abstract

Single-country studies document varying time trends in memory function and impairment. Comparative analyses are limited. We used self-respondent data on adults aged 50+ years in 13 countries from three surveys (USA: HRS, 1998-2018; England: ELSA, 2002-2018; 11 European countries: SHARE, 2004-2019). Memory is measured with tests of immediate and delayed word recall. Unweighted age- and gender-adjusted mixed effects regression models as well as models with adjustments for additional socio-demographic characteristics and health behaviors were examined. Heterogeneity in trends by gender, age group, and educational attainment were measured. The age-adjusted 10-year improvement in average test score is 0.04 standard deviations (SDs) (95% confidence interval (CI): 0.03, 0.05) in the USA, 0.17 SDs (95% CI: 0.15, 0.19) in England, and 0.24 SDs (95% CI: 0.23, 0.25) in SHARE countries. Trends are largely similar across gender, age groups, and educational attainment. Regional differences in trends remain after adjustment for potential mechanisms. Difference between the USA and other countries is particularly large under aged 75 years compared to over aged 75 years. Pace of improvement in memory function varies strongly across countries. On average, the 11 European countries studied had the fastest improvement, followed by England. The trend in the USA indicates improvement, but at a much slower pace compared to that in England and other European countries. Uncovering the causes for the cross-country heterogeneity in time trends, and in particular the reasons for the comparatively poor performance of the USA, should be both a research and public health priority.